Beatrix Becker

Zusammenfassung

Abstract Efficient compositional models are required to simulate underground gas storage in porous formations where, for example, gas quality (such as purity) and loss of gas due to dissolution are of interest. We first extend the concept of vertical equilibrium (VE) to compositional flow, and derive a compositional VE model by vertical integration. Second, we present a hybrid model that couples the efficient compositional VE model to a compositional full-dimensional model. Subdomains, where the compositional VE model is valid, are identified during simulation based on a VE criterion that compares the vertical profiles of relative permeability at equilibrium to the ones simulated by the full-dimensional model. We demonstrate the applicability of the hybrid model by simulating hydrogen storage in a radially symmetric, heterogeneous porous aquifer. The hybrid model shows excellent adaptivity over space and time for different permeability values in the heterogeneous region, and compares well to the full-dimensional model while being computationally efficient, resulting in a runtime of roughly one-third of the full-dimensional model. Based on the results, we assume that for larger simulation scales, the efficiency of this new model will increase even more.

Zusammenfassung

Abstract Hydrogen is a promising alternative to carbon based energy carriers and may be stored in large quantities in subsurface storage deposits. This work assesses the impact of static (density and phase equilibria) and dynamic (viscosity and diffusion coefficients) properties on the pressure field during the injection and extraction of hydrogen in the porous subsurface. In a first step, we derive transport properties for water, hydrogen and their mixture using the Perturbed-Chain Statistical Associating Fluid Theory equation of state in combination with an entropy scaling approach and compare model predictions to alternative models from the literature. Our model compares excellently to experimental transport coefficients and models from literature with a higher number of adjustable parameters, such as GERG2008, and shows a clear improvement over empirical correlations for transport coefficients of hydrogen. In a second step, we determine the effect of further model reduction by comparing our against a much simpler model applying empirical transport coefficients from the literature. For this purpose, hydrogen is periodically injected into and extracted out of a dome-shaped porous aquifer under a caprock. Our results show that density and viscosity of hydrogen have the highest impact on the pressure field, and that a thermodynamic model like the new model presented here is essential for modeling the storage aquifer, while keeping the number of coefficients at a minimum. In diffusion-dominated settings such as the diffusion of hydrogen through the caprock, our developed diffusion coefficients show a much improved dependence on temperature and pressure, leading to a more accurate approximation of the diffusive fluxes.

Zusammenfassung

We present version 3 of the open-source simulator for flow and transport processes in porous media DuMux. DuMux is based on the modular C++ framework Dune (Distributed and Unified Numerics Environment) and is developed as a research code with a focus on modularity and reusability. We describe recent efforts in improving the transparency and efficiency of the development process and community-building, as well as efforts towards quality assurance and reproducible research. In addition to a major redesign of many simulation components in order to facilitate setting up complex simulations in DuMux, version 3 introduces a more consistent abstraction of finite volume schemes. Finally, the new framework for multi-domain simulations is described, and three numerical examples demonstrate its flexibility.